git-base-pokemon

This model is a fine-tuned version of microsoft/git-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0372
• Wer Score: 2.1641

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Score
7.3017	4.17	50	4.4850	20.9419
2.3174	8.33	100	0.4334	8.6357
0.1315	12.5	150	0.0334	0.4225
0.0175	16.67	200	0.0309	1.1072
0.0057	20.83	250	0.0330	2.3527
0.0024	25.0	300	0.0351	2.5284
0.0017	29.17	350	0.0358	2.4496
0.0014	33.33	400	0.0364	2.3941
0.0012	37.5	450	0.0367	2.1628
0.0011	41.67	500	0.0371	2.2390
0.0011	45.83	550	0.0372	2.1447
0.001	50.0	600	0.0372	2.1641

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3